LiveCode Forums.

I make a backup just about every time I save
my code.

With revAltArchive they have a string of numbers
added so that someprog.rev is saved as
someprog00001.rev.

When something goes wrong I open an appropriate
archive copy.

My confusion arrises because it opens as "someprog.rev".
As I've probably made further changes to "someprog.rev"
I don't want to save it with the original programme name.

This leads to all sorts of annoying things... "another copy
in memory" and in another case when I "close and
remove from memory" as well as the open archive
file there's an "untitled1.rev" which doesn't exist
on my computer.

This has caused me a lot of lost time and work.

Have any users figured out an efficient way of
making backups without all the problems I'm having?

Is there any info available which explains this unique
file system?

Why is it set up this way? Is it a Mac thing?

Starting to feel that it's time to give up on Rev.

Hi bjb,

as you have noticed, the filename and the name of the stack are 2 different properties of the stack. Basically what the archiving plugin does is not much more than a save as. So in the process the name property of the stack is not changed, while the filename is. Without having looked at altArchive, I suspect its scripts are open (or ping chipp to ask for permission to peek at the scripts). You could modify the script of the plugin in a way, that temporary renames the stack, before it does its save as, and then names it back to its original name, once the save is done. This is not too hard to do.

looking back at the main problem (another stack with the same name blahblah) I can only say I feel your pain. Really. This bothers me since I ever got started with rev and I would love to see that behaviour changed somehow. The reason is not a mac thing, but a design decision that has been made back in the day when the engine was still called Metacard. The rev IDE is trying to be helpful here. In theory it is possible to have two stacks of the same name in memory, but then the engine gets confused on which is which. So the IDE warns you that you might loose data, if you continue with the loading process. I really wish there was a way to refer to a stack by script, other then its name, but we only can wait and watch if that gets changed at some point.

All the best,

Malte

Thanks for the sympathetic reply malte.

Reminds me of the days when one of the
most popular operating systems was called
C/PM. The problems arose becaust the
C/PM OS as supplied to computer manufacturers
didn't specify a file format.

The result was that every computer maker
"rolled their own" so one had to have
handy a floppy with dozens (or was it
hundreds) of converters to use if you
had a friend with a computer and wanted
to exchange files.

This ended when IBM came into the PC market.
Everyone wanted to be compatible with them
so soon we had only one file system.

Enter left Bill Gates....

Yes, you are welcome to modify revAltArchive as you like.

The name conflict has zero to do with the filename of the stack. Name conflicts occur when two stacks have the same name-- even if their filenames are different. It's always a good idea to quit after you get such a message from the IDE.

You can always 'rename' a stack in the IDE and then there is no conflict.

-Chipp